阅读说明：本技术 用于处理语音无线电信号的装置、方法和计算机程序 (Apparatus, method and computer program for processing a speech radio signal ) 是由 奥莱·约翰 伊万·兹维列夫 于 2019-11-19 设计创作，主要内容包括：一种用于处理语音无线电信号的装置,其中转录单元被配置为将语音无线电信号转换为文本信号；对象确定单元,被配置为确定语音无线电信号源自的对象；对象定位单元,被配置为确定语音无线电信号源自的对象的位置信息；以及输出单元,被配置为将文本信号分配给对象并提供所述文本信号。对象确定单元(130)被配置为为其位置(210)至少部分地与所确定的位置信息(142)匹配的至少一个对象(200)确定检测概率(135、1351至1353)。对象确定单元(130)被配置为将具有最高检测概率(135、1351至1353)的对象(200)确定为语音无线电信号(110)源自的对象(200),或者,在具有非常相似检测概率的情况下,将所有具有相似检测概率的对象确定为所述对象。(An apparatus for processing a speech radio signal, wherein a transcription unit is configured to convert the speech radio signal into a text signal; an object determination unit configured to determine an object from which a voice radio signal originates; an object locating unit configured to determine position information of an object from which the voice radio signal originates; and an output unit configured to assign a text signal to an object and provide the text signal. The object determination unit (130) is configured to determine a detection probability (135, 1351 to 1353) for at least one object (200) whose location (210) at least partially matches the determined location information (142). The object determination unit (130) is configured to determine the object (200) having the highest detection probability (135, 1351 to 1353) as the object (200) from which the speech radio signal (110) originates or, in case of very similar detection probabilities, all objects having similar detection probabilities as the object.)

1. An apparatus (100) for processing a voice radio signal (110), comprising:

a transcription unit (120) configured to convert the speech radio signal (110) into a text signal (112);

an object determination unit (130) configured to determine an object (200) from which the speech radio signal (110) originates;

an object localization unit (140) configured to determine position information (142) of an object (200) from which the speech radio signal (110) originates;

an output unit (150) configured to assign the text signal (112) to the object (200) and to provide the text signal; and

wherein the object determination unit (130) is configured to determine a detection probability (135 ) for at least one object (200) whose position (210) at least partially matches the determined position information (142)₁To 135₃) And an

Wherein the object determination unit (130) is configured to

Will have the highest probability of detection (135 )₁To 135₃) Is determined as the object (200) from which the speech radio signal (110) originates, or

In case of having very similar detection probabilities, all objects having similar detection probabilities are determined as the object.

2. The apparatus (100) according to claim 1, wherein the object localization unit (140) is configured to determine a region (220) in which the objects are arranged with a probability (200) as the location information (142), and

wherein the object localization unit (140) comprises or is configured to communicate with at least one localization apparatus to determine the source of the speech radio signal as the area (220).

3. The apparatus (100) of claim 2, wherein the positioning device comprises at least one radio direction finder (144)₁To 144_n)。

4. The apparatus (100) according to one of claims 1 to 3, wherein the object localization unit (140) is further configured to receive position data (142 b) of an object (200)₁To 142b₅)。

5. The apparatus (100) of claim 4, wherein the object positioning unit (140) comprises or is configured to communicate with an AIS receiver (147), an ADS-B receiver (146), a radar unit and/or a general position data receiver (148) to receive position data (142B)₁To 142b₅) And an

Wherein the position data (142 b)₁To 142b₅) Including GPS position, course, speed, and/or altitude relative to sea level.

6. The apparatus (100) according to one of claims 1 to 5, wherein the object determination unit (130) comprises an AIS receiver (147), an ADS-B receiver (146) and/or a generic object identification receiver or is configured to communicate with the aforementioned receiver to obtain object identification data (132) of at least one object (200) whose position (210) at least partially matches the position information (142) determined by the object positioning unit (140).

7. The apparatus (100) of claim 6, wherein the object identification data (132) comprises a calling number of a Mobile services at sea (MMSI), an object name, a target of the object (200), a load of the object (200), and/or a size of the object (200).

8. The apparatus (100) of claim 1, wherein the detection probability (135 ) is a function of a signal from a signal source₁To 135₃) Defining a degree of correspondence of the determined position information (142) with an actual position (210) of the object (200), and/or

Wherein the object determination unit (130) is configured to determine the detection probability (135 ) based on a probability of correct position information (142) of the object positioning unit (140)₁To 135₃)。

9. The apparatus (100) according to one of claims 1 to 8, wherein the object determination unit (130) is configured to communicate with the transcription unit (120) to determine object identification data (132) of the object (200) from a text signal (112).

10. The apparatus (100) according to one of claims 1 to 9, wherein the transcription unit (120) is configured to extract a speech mode code (122) from the speech radio signal (110) and provide it to the object determination unit (130),

wherein the object determination unit (130) is configured to determine an object (200) from which the speech radio signal (110) originates based on the speech mode code (122).

11. The apparatus (100) according to one of claims 1 to 10, wherein the transcription unit (120) is configured to convert the speech radio signal (110) into a text signal (112) using a neural network.

12. The apparatus (100) according to one of claims 1 to 11, wherein the apparatus (100) is configured to process at least two voice radio signals (110) simultaneously and/or offset in time, and

wherein the output unit (150) is configured to assign at least two text signals (112) of the at least two speech radio signals (110) to respective objects (200) and to provide them chronologically to the apparatus (100) via a user interface (155) and/or to store them in a database.

13. The apparatus (100) according to one of claims 1 to 12, wherein the output unit (150) is configured to provide a text signal (112), the assigned object (200), the location (210) of the object (200) and the input time of the voice radio signal to the apparatus (100) via a user interface (155) and/or to store it in a database.

14. The apparatus (100) according to one of claims 1 to 13, wherein the object (200) is a ship, an airplane or a vehicle.

15. A method (1000) for processing a voice radio signal, the method comprising the steps of:

-converting (1100) the speech radio signal into a text signal by means of a transcription unit;

determining (1200), by an object determination unit, an object from which the speech radio signal originates;

determining (1300), by an object locating unit, position information of an object from which the speech radio signal originates; and

assigning (1400) the text signal to the object by an output unit and providing (1500) the text signal assigned to the object;

wherein determining (1200) the object comprises

Determining a detection probability for at least one object whose position at least partially matches the determined position information, an

The object having the highest detection probability is determined as the object from which the speech radio signal originates.

16. Computer program having a program code for performing the method according to claim 15 when the program runs on a computer.

Technical Field

Embodiments according to the invention relate to an apparatus, a method and a computer program for processing a speech radio signal.

Background

In the maritime, aviation and land-based sectors, there are currently no technical solutions that allow voice radio tracking (e.g., VHF marine radio, aviation radio, VHF terrestrial radio, etc.) connected with transmitter identification. At present, systems are not known that combine voice identification, transmitter information data evaluation, such as AIS data (marine) or ADS-B data (aeronautical), and radial direction finding techniques.

Today, many versions of radio equipment have a recording function, which can store received voice radio signals for a defined period of time, so that the same content can be played back later (last call voice recording). Therefore, today, only a fragmentary portion of the communication history can be played back as a recording for a short period of time. Furthermore, no identification of the sender of the radio message or assignment of the received communication to radio stations located in the receiving area (e.g. ship, airplane, land vehicle, etc.) is given. The sender of the radio information can only be determined indirectly through existing information systems, such as AIS, ADS-B, GPS data, etc.

In view of this, there is a need for a concept that allows for improved intelligibility of received voice radio signals, traceability through past and possibly missed voice radio signals, and location and identification of senders.

This object is solved by the independent claims with apparatus claim 1, method claim 16 and computer program claim 17.

Further developments of the invention are defined in the dependent claims.

Summary of The Invention

Embodiments relate to an apparatus for processing a speech radio signal, comprising a transcription unit configured to convert the speech radio signal into a text signal. Further, the apparatus comprises an object determination unit configured to determine an object from which the speech radio signal originates. Further, the apparatus comprises an object localization unit configured to determine position information of an object from which the speech radio signal originates; and an output unit configured to assign a text signal to the object and provide the text signal. The voice radio signal may be a voice message transmitted by a radio signal transmitter to a radio signal receiver, wherein the object may include the radio signal transmitter and the radio signal receiver. The text signal determined by the transcription unit may represent the speech radio signal as a message in text form (e.g., ASCII).

This embodiment of the device is based on the following findings: when converting a speech radio signal into a text signal by means of a transcription unit, the communication via radio is easy to track, since the information of the radio communication can be determined at any time in the text signal, so that erroneous memorization of the speech radio communication can be prevented. Furthermore, the assignment of position information of the object and/or object identification information to the text signal allows to identify or locate the sender of the voice radio signal, so that radio communication can be recorded very accurately and well. In particular, for example in rescue operations in land, air or sea areas, it is advantageous to be able to track and assign radio communications to individual objects during and after operation, such as ships, aircraft or land vehicles. Each object that may transmit voice radio signals may also include an apparatus as described herein. Thus, the device can allow each object participating in radio communication to track by the device which object transmitted the text signal recorded by the device as a voice radio signal when and where.

It must therefore be stated that this device is configured to render the voice radio signal understandable by converting it into a text signal, to read or study the content from the voice radio signal by means of the text signal, and to identify and locate the sender of the voice radio signal.

According to an embodiment, the object localization unit is configured to determine, with probability, an area in which the object is located as the location information. The object locating may comprise or be configured to communicate with at least one locating device to determine the source of the voice radio signal as a region. For example, this region is an extension in one dimension (e.g., a line indicating the direction of an incoming voice radio signal (e.g., a signal beam)), two dimensions (a region of any form, such as a circular region, a circular sector, a triangle, a rectangle, a polygon, etc.), or three dimensions (e.g., a body of any shape, such as a spherical region, a conical region, a cuboid region, etc.). For example, this area defines the direction from which the speech radio signal originates, from which direction the object localization unit can infer that the object is arranged in this area. Here, the object localization unit may determine a probability that the object is arranged in the region, wherein the probability may indicate how accurately the at least one localization apparatus may determine the region. Thus, a coarse positioning has become possible, whereby this device can be used for e.g. rescue actions, since even with poor speech intelligibility (e.g. the location of the object is e.g. unintelligible or not communicated) the object sending out the emergency radio signal can be positioned by the object positioning unit in order to dispatch a rescue effort in the direction of the object (e.g. this area).

According to an embodiment, the positioning device comprises at least one radio direction finder. The object-locating device may confine the source of the speech radio signals within this area by means of at least one radio direction finder. In other words, this area may include the source of the voice radio signal. If, for example, a plurality of radio direction finders is used, the area may be reduced and/or the probability may be increased or the exact position of the object may be determined by the object localization unit.

According to an embodiment, the object location unit is further configured to receive location data (e.g. GPS location, route, speed, etc.) of the object. For example, here, rather than a voice radio signal being located, an object is disposed within a radius of the apparatus (e.g., a maximum distance of 20 kilometers, 50 kilometers, 100 kilometers, or 1000 kilometers from the apparatus). This allows determining position information of potential objects from which the speech radio signal may originate. Optionally, the object locating unit is configured to receive position data (e.g. GPS position, route, speed, etc.) of the object in the area (the area from which the voice radio signal originates). This allows not only to determine the area from which the voice radio signal originates (e.g. by a positioning device), but also to determine very accurate position data of objects which may have emitted the voice radio signal. Therefore, the positioning of the object that has emitted the voice radio signal (or the object determination by the object determination unit) is not limited to the entire area but individual positions within the area. This allows for an optimized positioning and allocation of the voice radio signals of the objects.

According to embodiments, the object locating unit may include or may communicate with an AIS receiver, an ADS-B receiver, a radar unit, and/or a general location data receiver to receive location data. The location data may include GPS location, course, speed, and/or altitude relative to sea level. Here, the position data may be received by different receivers according to the object. Thus, for example, an AIS receiver may receive position data of a ship, an ADS-B receiver may receive position data of an airplane, a radar unit may receive position data of a metallic object, such as a ship, an airplane, a vehicle, etc., and a general position data receiver may receive position data of a plurality of possible objects, such as a land vehicle. This allows the device for processing voice radio signals to be used many times on land, in water and in the air. The specific combination of a receiver for receiving position data of an object and a radio locator determines with probability in the object locating unit the region from which the voice radio signal originates, allowing to determine the object from which the voice radio signal originates in a very fast and accurate manner. Thus, the voice radio signal can be very efficiently distributed to the object and its position.

According to an embodiment, the object determination unit may comprise an AIS receiver, an ADS-B receiver, and/or a generic object identification receiver, or may communicate therewith to obtain object identification data of at least one object whose position at least partially matches the position information determined by the object positioning unit. Such matching may mean, for example, that the object determination unit compares the position data of the object with the region comprising the source of the voice radio signal and receives only the object identification data of the object in the region. Thus, the object determination unit is configured to assign object identification data, such as a calling number of a mobile services at sea (MMSI), an object name, a target of the object, a load of the object and/or a size of the object positioned with respect to the object positioning unit, for example. Thus, the voice radio signal can be assigned by the device not only to the position information, for example the localization of the voice radio signal, but additionally to the object from which the voice radio signal originates. Here, the object determination unit is configured to obtain object identification data of a ship, for example via an AIS receiver, of an airplane via an ADS-B receiver, and/or of at least one object via a general object identification receiver of other objects, such as a land vehicle. Here, the position information determined by the object positioning unit may represent, for example, an accurate GPS position, whereby the object determination unit acquires, for example, object identification data only from an object whose position exactly matches the determined position information. For example, when the position information determined by the object positioning unit defines an area, the object determining unit may acquire the object identification data from a plurality of objects whose positions are within the area.

According to an embodiment, the object identification data may include a calling number of a mobile services at sea (MMSI), an object name, a target of the object, a load of the object, and/or a size of the object. Thus, by assigning a calling number for a marine mobile service, the user of the device can very easily contact the object from which e.g. a voice radio signal originates. Further, in radio communication of a plurality of objects, by assigning object names to corresponding voice radio signals, it is possible to assign different voice radio signals to the respective objects by the object names to improve traceability of the radio communication. The object's target, the object's load and/or the object's size may represent further important object identification data which, together with the speech radio signal, may represent very detailed information that needs to be processed efficiently in radio communication. Here, it is particularly advantageous if the apparatus is configured to provide the speech radio signal in the form of a text signal together with the object identification data (e.g. via the output unit).

According to an embodiment, the object determination unit is configured to determine a detection probability of the at least one object whose position at least partially matches the determined position information. Further, the object determination unit is configured to determine, for example, the object having the highest detection probability as the object from which the voice radio signal originates. For example, the detection probability defines the probability of an object from which a speech radio signal originates. For example, when the object determination unit identifies several objects, for example, whose positions at least partially match the position information determined by the object localization unit, the detection probability allows the object determination unit to assign a single object to a speech radio signal or to a corresponding text signal. Thus, the device can unambiguously assign the object to a speech radio signal or a text signal.

According to an embodiment, in case of very similar detection probabilities (e.g. a deviation of ± 1%, ± 2% or ± 5%), the object determination unit may be configured to determine all objects with similar detection probabilities as said objects, and for example, the output unit may be configured to assign all these objects to the text signal in that case and also to indicate the respective detection probabilities. Optionally, in that case, the apparatus may be configured to analyze at least two voice radio signals from the same subject following each other in rapid succession (e.g. within at most 5 minutes, within at most 30 minutes, within at most 1 hour or within at most 5 hours) to increase the detection probability. The position of the object may have changed between the at least two voice radio signals and this change of position may be compared, for example, by the object determination unit, to a route or course of the object whose position at least partly matches the position information. According to an embodiment, the apparatus may be configured to determine, by the speech mode code, whether at least two speech radio signals following each other in rapid succession originate from the same object.

According to an embodiment, the detection probability defines a degree of correspondence of the determined position information with an actual position of the object, wherein the determined position information represents position information determined by the object localization unit. Additionally or alternatively, the object determination unit may determine the detection probability based on a probability of correct position information of the object localization unit. The determined position information may represent, for example, a region that has been determined by the object localization unit, for example by means of a radio direction finder, and the actual position of the object may be arranged, for example, at the edge of the region or in the center of the region or only partially overlapping the region, which results in different degrees of correspondence defining the detection probability. Thus, for example, an object near the center of a region has a higher detection probability than an object located at the edge of the region. The additional use of the probability of correct position information determined by the object localization unit additionally or alternatively allows incorporating possible inaccuracies of the device and thus allows very accurate identification of the object. Here, the probability of correct position information may correspond to, for example, the probability with which the object positioning unit is assigned to the region, the probability indicating in which probability the object is arranged in the region determined by the object positioning unit.

According to an embodiment, the object determination unit is configured to communicate with the transcription unit to determine an object identification of the object from the text signal. Thus, for example, the speech radio signal may already comprise an object identification (in this way, the radio operator transmitting the speech radio signal may state his name and/or the name or identification of the object transmitting the speech radio signal), which may be transcribed by the transcription unit and may be determined from the transcribed text signal by the object determination unit. This allows for a determined, e.g. a hundred percent or a very high, detection probability of the object by the object determination unit without having to compare the position of the object with the position information determined by the object localization unit. Alternatively, for verification, a comparison may still be made.

According to an embodiment, the transcription unit is configured to extract the speech pattern code from the speech radio signal and provide it to the object determination unit. Furthermore, the object determination unit may be configured to determine the object from which the speech radio signal originates based on the speech mode code. The speech mode code may be assigned to, for example, a radio operator who may be assigned to the subject. According to an embodiment, the apparatus may comprise or may be configured to access a database, wherein the database may comprise a speech pattern code assigned to a radio operator or object. Alternatively, the apparatus may also extract the speech pattern code from the first speech radio signal, determine the assigned object by the object determination unit, and then cache the speech pattern code together with the object to detect the speech pattern code in the following second speech radio signal without having to determine the object again by the object determination unit, but determine the assigned object identification information directly from the cached speech pattern code. In other words, the object determination unit may be configured to first determine the object identification data of the object independently of the determined speech pattern code and to determine the object identification data of the object based on the speech pattern code in the second speech radio signal having the same speech pattern code.

According to an embodiment, the transcription unit is configured to convert the speech radio signal into a text signal using a neural network. This allows, for example, the transcription unit to detect commonly used phrases in speech radio signals through a neural network, thus enabling a very efficient and easy conversion of speech radio signals into text signals.

According to an embodiment, the apparatus is configured to process at least two voice radio signals simultaneously and/or offset in time. Furthermore, the output unit may be configured to assign the at least two text signals to the at least two speech radio signals of the corresponding object and to provide them to the device in chronological order via the user interface and/or to store them in the database. This allows tracking of the radio communication path using a plurality of voice radio signals or studying earlier voice radio signals and assigning them to corresponding objects. Thus, the apparatus may be configured to record and provide a file record of radio communications having a plurality of voice radio signals.

According to an embodiment, the output unit is configured to provide the text signal, the assigned object, the position of the object and the input time or voice radio signal to the apparatus via the user interface and/or to store these in a database. Here, for example, the output unit may be configured to provide a text signal with the assigned object in the position as text data or to store this text data, wherein the data may be indicated on a user interface, such as in a chat log, for example. Furthermore, card material such as land, water or air may also be displayed on the user interface and the object is indicated with a text signal at the position determined by the object localization unit. Thus, on the user interface, the text signal can be illustrated with the assigned object, for example, by the output unit. Both the user interface and the database enable fast access to data determined by the device.

According to an embodiment, the object is a ship, an airplane or a vehicle.

An embodiment provides a method of processing a speech radio signal, wherein the method comprises converting the speech radio signal into a text signal by a transcription unit, determining an object from which the speech radio signal originates by an object determination unit, determining position information of the object from which the speech radio signal originates by an object location unit, and assigning the text signal to the object, and providing the text signal assigned to the object by an output unit.

Embodiments provide a computer program having a program code for performing the method described herein, when the program is run on a computer.

Drawings

Embodiments according to the invention will be discussed in more detail below with reference to the accompanying drawings. With regard to the illustrated schematic drawings, it should be noted that the illustrated functional blocks are to be considered implementations or features of the inventive apparatus and method steps of the inventive method and method steps from which the inventive method can be derived. They show that:

FIG. 1 is a schematic illustration of an apparatus according to an embodiment of the invention;

FIG. 2 is a schematic block diagram of an apparatus according to an embodiment of the present invention;

fig. 3 is a schematic diagram of graphic provision of a text signal assigned to an object through an output unit according to an embodiment of the present invention;

fig. 4 is a schematic diagram of a graphic provision of a text signal assigned to an object having the highest detection probability by an output unit according to an embodiment of the present invention;

FIG. 5 is a schematic illustration of non-unique identification of an object emitting a voice radio signal by an apparatus according to an embodiment of the present invention; and

FIG. 6 is a block diagram of a method of processing a voice radio signal according to an embodiment of the present invention;

Detailed Description

Before embodiments of the invention are discussed in more detail below with reference to the drawings, it should be noted that identical, functionally identical, or identical elements, objects, and/or structures have the same or similar reference numerals in the different drawings so that the descriptions of these elements illustrated in the different embodiments are interchangeable or mutually applicable.

Fig. 1 shows a schematic diagram of an apparatus 100 for processing a voice radio signal 110. The apparatus 100 comprises a transcription unit 120 configured to convert the speech radio signal 110 into a text signal 112. Furthermore, the apparatus 100 comprises an object determination unit 130 configured to determine an object 200 from which the speech radio signal 110 originates. Furthermore, the apparatus 100 comprises an object localization unit 140 configured to determine position information 142 of the object 200 from which the speech radio signal 110 originates, and the apparatus 100 further comprises an output unit 150 configured to assign the text signal 112 to the object 200 and to provide the text signal.

According to an embodiment, the object localization unit 140 may be configured to determine the region 220 as the location information 142 to which the object 200 is assigned with a probability, wherein the probability may indicate the accuracy with which the object localization unit 140 determines the location information. According to fig. 1, the region 220 may have a three-dimensional extension. However, the region 220 may also have a two-dimensional (e.g., area) or one-dimensional (e.g., signal beam) extension. To determine the location information 142, the object-locating unit 140 may include or may be configured to communicate with at least one locating device to determine an area 220 including a voice radio signal source or to determine an accurate location 210 of an object.

According to an embodiment, the positioning device may comprise at least one radio direction finder. Here, it should be noted that when the object positioning unit 140 includes or communicates with only one radio direction finder, the object positioning unit 140 may be configured to determine the region 220 as the position information 142. If the object-locating unit 140 includes or is configured to communicate with at least two radio-direction finders, a high approximation to the exact position 210 may be determined by taking into account system inaccuracies, since triangulation is possible in this case.

According to an embodiment, the object localization unit 140 may further be configured to receive location data, such as an accurate location 210 of the object, from the object 200 in the area 220. Thus, the apparatus 100 may be configured to first determine the area 220 and subsequently determine the position data 210 of the object 200 in the area and determine or provide the position data 210 as the position information 142 instead of the area 220.

According to an embodiment, the object localization unit 140 may be configured to detect whether an area or an accurate position of the object 200 may be determined with the radio direction finder and to decide on whether the position data 210 of the object 200 is to be detected in the area 220 or not accordingly. Here, the object localization unit 140 may be configured to receive the position data 210 when only one area is determined using one or more radio scanners, or may decide not to receive the position data 210 when the object localization unit 140 has determined the position data 210 by several radio scanners.

According to an embodiment, the object positioning unit 140 may include, or may be configured to communicate with, an AES receiver, an ADS-B receiver, a radar unit, and/or a general location data receiver to receive the location data 210. As shown in fig. 1, the location data 210 may include, for example, GPS location and additionally or alternatively routes, speeds, and/or altitudes relative to sea level. Here, the list of possible location data receivers and location data 210 is considered to be exemplary and not limiting.

In the following, the features and functions of the object localization unit 140 will be discussed in other words, wherein in particular ships and airplanes are treated as objects 200 to be localized.

The ship can be located as well as the airplane can be located by the object locating unit 140 having different technologies. AIS technology (positioning ship), ADS-B technology (airplane), radar equipment, and radio direction finders are all suitable systems.

AIS stands for automatic identification system (ETSI EN 303098-1, 2013-05, page 9). Generally, AIS systems allow, for example, monitoring of marine traffic and for preventing collisions between ships. The functional basis of the system is a ship facility equipped with AIS transponders and electronic sea maps (ECDIS). In the time division multiple access method (TDMA), the transponders transmit data, for example frequencies (161.975MHz and 162.025MHz) that can be received from other ship or land stations (the international telecommunication union, pages 1 and 67) (also via AIS transponders). The data includes, for example, information such as ship identification, current GPS location, route, speed, and other ship-related data. For example, the data may be displayed on an electronic chart. When the ship is equipped with AIS, they are correspondingly visible to others, as well as others. The object-locating unit 140 is configured to receive and process, for example, these data to determine location information 142.

ADS-B stands for broadcast auto-correlation monitoring (see specification in EUROAE ED 129 "TECHNICAL SPECIFICATION FOR A1090 MHZ EXTENDED SQUITTER ADS-B GROUND SYSTEM") and allows, similar to AIS SYSTEMs in the maritime domain, the acquisition of aircraft within the range of the relevant ADS-B receiver. When the aircraft are equipped with ADS-B transponders, they will transmit their identification, route, speed, current location, and other data over, for example, 1090 MHz. Thus, other aircraft and air traffic controllers may see the aircraft. The object-locating unit 140 is configured to receive and process, for example, these data to determine location information 142.

Unlike the AIS system and ADS-B system, the radar unit does not rely on the interchange of data. The basis for radar localization is the reflectivity of the conductive surface. Most ships, airplanes, and land vehicles (taking object 200 as an example) have a metal body that reflects incident electromagnetic waves. Thus, the radar can send high frequency transmit pulses and can subsequently receive echoes. Since the propagation speed of the radio wave is known (speed of light), the distance from the radar station to the ship can be determined from the time measurement between the transmitted pulse and the echo. In order to determine the relative angle of the echo signals, a mechanical rotation antenna is often used to transmit high-frequency pulses in various directions and receive echoes from the same direction.

Both AIS systems and radar systems allow for the display of charts of multiple ships within the corresponding range of the unit. In the context of communication security, AIS, ADS-B, and radar localization determine only the number of potential communication participants.

The radio direction finding system measures the incident angle of the electromagnetic wave on the direction finding antenna to determine the direction of origin of the radio signal. Basically, direction-finding systems "analyze" the electromagnetic wave field around a direction-finding antenna that is composed mainly of several dipole elements. For this reason, different direction finding methods exist. Interferometers and doppler systems are commonly used. The interferometer principle uses direct measurement of the phase difference between the individual elements of the direction-finding antenna. Knowing the distance between the antenna elements and the speed of light allows the azimuth to be calculated geometrically. Interferometer direction finders typically require one receiver per antenna element. Such systems allow direction finding of very short radio pulses, which is very important for radio monitoring, compared to doppler systems. In the doppler principle, the individual antenna radiators of a direction finding antenna are switched (commutated) so that the direction, when the antenna is found, ultimately represents a uniform motion of the virtual antenna vibrator on a circular orbit in the incident electromagnetic wave. If the co-directional wave moves, the reception frequency will increase according to the doppler. When the virtual antenna is far away from the wave, the reception frequency may decrease. This results in a frequency modulation at the receiver input which is demodulated and processed during signal processing as a "direction-finding phase signal". If the incident direction of the radio signal changes, the phase position of the direction-finding phase signal also changes. An azimuth angle is determined by measuring the phase position. Since both the relative angle and the distance to the communication source should be known locations, no single direction-finding system should be used for location determination. Accurate or very accurate radio positioning can be performed by cross-positioning. However, this requires a further direction finding system at a significantly distant installation location. For space reasons, it is almost impossible to use a radio positioning system by cross positioning method on a ship or an airplane. However, in land-based applications, such as sea-road control or flight control, multiple radio direction-finding systems located at different locations may be used.

According to an embodiment, the object determination unit 130 may further comprise an AIS receiver, an ADS-B receiver, and/or a generic object identification receiver, or may be configured to communicate therewith to obtain object identification data 132 of the at least one object 200, the location 210 of which at least partially matches the location information 142 determined by the object positioning unit 140. Here, the object determination unit 130 and the object location unit 140 may share the same AIS receiver, ADS-B receiver, or both the object determination unit 130 and the object location unit 140 are configured to communicate with the same AIS receiver or ADS-B receiver. Furthermore, the object localization unit 140 may be configured to provide the object determination unit 130 with the position information 142, such that the object determination unit 130 may compare the position 210 of the object 200 with the position information 142. It is thus possible that the object determination unit 130 determines, for example, only the object identification data 132 of the object 200 which at least partially comprises the position information 142, thereby ensuring that the object 200 transmits the voice radio signal 110 with a high probability.

According to an embodiment, the object identification data 132 includes a calling number of a Mobile services at sea (MMSI), an object name, a target of the object, a load of the object, and/or a size of the object. The list of object identification data is considered here as exemplary and not limiting.

According to an embodiment, the object determination unit 130 is configured to determine a detection probability of the at least one object 200, the location 210 of which at least partially matches the determined location information 142. The detection probability may define with what probability or certainty a certain object 200 has sent the voice radio signal 110, for example. Further, the object determination unit 130 may be configured to determine the object having the highest detection probability as the object 200 from which the voice radio signal 110 originates. Thus, for example, only the data of the object 200 having the highest detection probability is determined as the object identification data 132.

According to an embodiment, the detection probability determines how much the determined position information 142 corresponds to the actual position 210 of the object 200. The closer the object 200 is arranged, e.g. closer to the determined position information 142, the higher the detection probability. Additionally or alternatively, the object determination unit 130 may be configured to determine the detection probability based on the probability of the correct position information 142 of the object localization unit 140. Thereby, the object determination unit 130 may incorporate uncertainty or possible errors of the object localization unit 140 in determining the object 200.

According to an embodiment, the object determination unit 130 may be configured to communicate with the transcription unit 120 to determine the object identification 132 of the object 200 from the text signal 112.

According to an embodiment, the transcription unit 120 is configured to extract the speech pattern code 122 from the speech radio signal 110 and provide it to the object determination unit 130, wherein the object determination unit 130 may be configured to determine the object 200 from which the speech radio signal 110 originates based on the speech pattern code 122. Here, the object determination unit 130 may associate, for example, a person on a ship, an airplane, or a land vehicle with the voice pattern code 122 of the corresponding object (e.g., ship, airplane, or land vehicle) to determine the object. According to an embodiment, the object determination unit 130 may provide the object determined in this manner as the object identification data 132.

According to an embodiment, the transcription unit 120 may be configured to convert the speech radio signal 110 into the text signal 112 using a neural network. The apparatus 100 therefore comprises an advantageous transcription unit 120, since the radio speech signal 110 can be converted very quickly into a text signal 112 by a neural network.

According to an embodiment, the transcription unit 120 may comprise an already existing speech processing system for converting a voice message (e.g., the voice radio signal 110) into a text signal 112. Accordingly, transcription unit 120 may include known speech recognition software, such as described briefly below. Automatic identification and understanding of spoken language by computers has been the subject of intense research for decades. Automatic speech recognition is a method that allows a computer to automatically detect spoken language as data and then make it available for processing. Currently, voice recognition software for voice processing and independent voice recognition is provided by a number of vendors and is in use.

According to an embodiment, the apparatus 100 is configured to process at least two voice radio signals 110 simultaneously and/or offset in time. Here, the at least two speech radio signals 110 may originate from different objects in different locations. Furthermore, the output unit 150 may be configured to assign at least two text signals 112 of the at least two speech radio signals 110 determined by the transcription unit 110 to the respective objects 200 and to provide and/or store them in a time sequence in a database via a user interface of the apparatus 100. Thus, the apparatus 100 is for example configured to record radio communications with at least two voice radio signals 110 in a traceable manner.

According to an embodiment, the output unit 150 is configured to provide and/or store the text signal 112, the assigned object 200, the location 210 of the object 200 and the input time of the voice radio signal 110 in a database via a user interface of the apparatus 100. Here, the output unit 150 may receive the text signal 112 from the transcription unit 120, the position of the object from the object localization unit 140 via the position information 142, and the assigned object via the object determination unit 130, for example, by means of the object identification data 132. The output unit 150 may be configured to process the text signal, the assigned object, the location of the object, and the input time of the voice radio signal 110 so that a user of the apparatus 100 may very easily and efficiently track or study the history of the radio communication.

According to an embodiment, the object 200 may be a ship, an airplane, or a vehicle.

According to an embodiment, the apparatus is configured to comprise at least one of the following three points:

in speech recognition, the apparatus 100 may comprise a programmed deep neural network for marine speech recognition,

this network has been trained or can be trained by the device.

In the identification of ships or of aircraft or land vehicles, the apparatus 100 may comprise developed algorithms, for example in an object identification unit 130, which identifies and locates one or more objects based on the input data 112, 122 and/or 142 (see fig. 2 "system diagram-block diagram").

Concatenation of speech identity and object identity.

In other words, the following embodiments will illustrate the context between identification and positioning. Four application cases are also involved:

a. complete-equipment ship

AIS system and one radio on board

b. Ship with simple equipment

-only one radio device

c. Complete-equipment airplane

ADS-B repeater system and a radio device on board

d. Aircraft with crude equipment

-only one radio on board

a. Complete-equipment ship

Such as an onboard AIS system and a radio.

Scene

For example, the captain reports over the radio on channel 16. The AIS transponder, for example, continuously transmits corresponding ship information (MMSI, ship name, location, speed, route, and other data).

Positioning

Positioning by a radio direction finder:

when the captain speaks, the direction of the radio signal is found by the device 100. Here, for example, the direction from the direction finding station to the ship is determined. In the sense of positioning, by knowing the radio direction finding deviation of the direction finding system, the position where the ship object 200 is, for example, a cone (e.g. area 220) becomes known. The processing algorithm of device 100 registers this cone as a region with increased probability of detection. Furthermore, evaluating the radio signal level has an influence on the probability distribution of the signal beam (e.g. area 220).

Furthermore, if direction finding is to be performed from different locations using a further direction finding system, a further "probability cone" is generated. Both probability regions are processed by an algorithm, which results in a limited region (e.g., region 220) having an increased probability of localization (e.g., probability of detection). Here, it is obvious that the basic positioning can already be performed by a direction-finding system. Evaluating the radio signal level and using a further direction finding system may improve the positioning accuracy.

Here, it can be said that object localization has occurred. Finally, an area (e.g., area 220) from which a radio message has been sent is known.

Positioning accuracy is improved by evaluating AIS data:

from the received AIS data, for example, position data (GPS position 210, route, speed) and identification data (MMSI, ship name, destination port, load, ship size, etc.) of the ship within the reception range of the unit are obtained. By measuring the time between the current time and the AIS message time, the current ship position can be determined more accurately while taking into account the ship's route and the ship's speed.

If one or several ships (objects 200) are in an already determined probability region (e.g. region 220, a region with a GPS position and an assigned detection probability), the ship position 210 with the highest probability will be detected as a radio signal source. The GPS position is obtained from the AIS data and corrected to terminate the position fix with the greatest possible system accuracy.

Identification

For example, the identification is derived from the positioning. All relevant identification data, such as MMSI, ship name, destination port, load, ship size, etc., is obtained from the assigned AIS message including the detected GPS location 210.

Transcription

For example, after receiving a voice signal, transcription is performed locally and automatically by transcription unit 120 based on a voice message (e.g., voice radio signal 110) transmitted via VHF marine radio. For example, a neural network is used for this purpose, which network is developed specifically for detecting standard maritime communication phrases. By linking the transcription system 120 to sender location (e.g., object location unit 140) and identification (e.g., object determination unit 130), received voice messages can be retrieved in written form (e.g., text signal 112) and can be assigned to a corresponding located ship so that past radio messages (e.g., voice radio signal 110) can be tracked via a user interface. If the transcribed voice message (e.g., text signal 112) includes an erroneous or undetected voice message, subsequent corrections can be made via a feedback loop so that the detection rate of the deep neuron network can be additionally optimized over time.

b. Ship with simple equipment

For example, there is only one radio on board.

Scene

For example, the captain reports over the radio on channel 16. Since this ship does not have, for example, an AIS transponder, corresponding ship information (MMSI, ship name, location, speed, route and other data) is not issued.

Positioning

For example, positioning by radio direction finding and evaluating signal strength is the same way as positioning a fully equipped ship. Since the ship does not transmit IIS data, there is a possibility that the detection probability of ship objects not within the determined probability area or other ships around is evaluated too low to determine a unique GPS position. Thus, the accuracy of locating the ship without the AIS transponder is low, by contrast. Furthermore, false detections may even occur when the transmitted GPS position is evaluated as being within a certain probability zone for a highly equipped ship.

Identification

In such a scenario, for example, identification may not necessarily be performed in an automated manner. It may be assumed that the radio signal is from a ship that does not need to be equipped with AIS equipment or that the AIS system is defective or shut down.

Transcription

The transcription function is the same as on a fully equipped ship, since transcription is only done locally based on the received VHF radio and therefore independent of the ship equipment that sent the voice message.

c. Complete-equipment airplane

For example, there is an ADS-B repeater system and a radio on board.

Scene

The pilot reports each radio, for example, at the known tower frequency (118-. ADS-B transponders constantly send corresponding information (identification, location, speed, route and other data).

Positioning

Positioning of the radio direction finder:

when the pilot speaks, direction finding of the radio signal is performed. Here, for example, the direction from the direction finding station to the airplane is determined. In the sense of positioning, by knowing the direction-finding deviation of the direction-finding system, for example the cone, the position of the aircraft can be known. The processing algorithm registers this cone as a region with increased (e.g., voice radio signal 110) (region 220) detection probability. Furthermore, the evaluation of the radio signal level has an influence on the probability distribution over the signal beam (area 220).

Furthermore, if direction finding is to be performed from different locations using a further direction finding system, a further "probability cone" is generated. For example, two probability regions are algorithmically processed, which results in a limited region (region 220) with increased probability of localization. Here, it is apparent that a direction-finding system has been used for preliminary positioning. Evaluating the radio signal level and using a further direction finding system may improve the positioning accuracy.

Here, it can be said that object localization has occurred. Finally, the area where the radio message is sent is known.

Positioning accuracy is improved by evaluating ADS-B data:

from the received ADS-B data, position data 210(GPS position, route, speed) and identification data (identification, aircraft type, etc.) of the aircraft (object 200) within the reception range of the unit are obtained. By measuring the time between the current time and the time of the ADS-B message, the current position of the aircraft may be more accurately determined taking into account the route and speed.

If one or more aircraft are within the determined probability region (field with GPS location and assigned detection probability), the aircraft location with the highest probability will be detected, e.g., as a radio signal source. The GPS position is obtained from ADS-B data and corrected to terminate positioning with the greatest possible system accuracy.

Identification

For example, the identification is derived from the positioning. For example, all relevant identification data, such as identification, airplane type, and other data, is obtained from the ADS-B message including the assignment of the detected GPS location.

d. Equipped with plain aircraft (e.g., UL-Ultralight)

For example, there is only one radio on board.

Scene

The pilot reports each radio, for example, at the known tower frequency (118-. For example, since the aircraft does not have an AIS transponder, corresponding information (identification, aircraft type, location, speed, route, and other data) is not sent out.

Positioning

The positioning takes place by radio direction finding and evaluation of the signal strength, for example in the same way as the positioning of a fully equipped aircraft. Since the aircraft or helicopter is not transmitting ADS-B data, the object may not be within the determined probability region or the detection probability of other aircraft is scored too low to determine a unique GPS location. In contrast, the accuracy of locating an aircraft without a transponder is low. Furthermore, fault detection may even occur when the transmitted GPS position is evaluated as being within a certain probability zone for a highly likely fully equipped aircraft.

Identification

In this context, for example, the identification may not necessarily be carried out in an automated manner. It may be assumed that the radio signal is from an aircraft that does not need to be equipped with ADS-B devices or that the transponder system is defective or turned off.

In land based uses, such as rescue services and disaster control, stationary or mobile (vehicular) mission control are equipped with the apparatus 100, in particular with the transcription unit 120, the object determination unit 130 and the object localization unit 140 (e.g. radio direction finder) to track radio messages (e.g. voice radio signals 110) from the deployed (in use) units. Thus, in the field of use of the marine department and the aviation department, evaluation of the situation in the mission control and recording of the situation can be similarly ensured.

The impact of the device 100 for automatic transcription of the speech radio signal 110 and for simultaneous identification of the sender and its location makes the communication in the radio more secure. The communicating participants (e.g., object 200) are supported because they clearly understand what is being spoken (speech recognition), the person who is speaking (identification), and the location where the object is located (location/positioning). By means of the technology, the traceability of complex communication structures in the maritime sector, air traffic and other application areas will be improved. An automatic transcription system (e.g., transcription unit 120) writes received radio communications locally and independently of the speaker and stores them in addition to linked transmitter detection, primarily to support and mitigate coastal radio stations, maritime search and rescue organizations, public authorities, and crew from accomplishing their tasks. Furthermore, this use supports marine training when using the ship guidance simulator. In the field of aviation, the system can be used for improving communication safety, relieving work of air traffic controllers and the like. Similar advantages can be determined for further fields of application.

Marine applications:

rescue organizations, such as DGzRS (maritime search and rescue services, germany) or havierkommando (central commander of maritime emergencies), would benefit greatly from secure communications during rescue operations. By identifying, locating and tracking the identity, location and tracking of the damaged ship emergency call, rescue actions can be organized more quickly and efficiently.

Water police, coastal police, VTS (ship traffic service) service providers and other organizations that function monitoring on behalf of their work-important aspects can also use the presented technology in an advantageous manner.

In the device 100 described herein, emphasis may also be placed on the integration of this technology with existing systems. A possible manufacturer of ECDIS (electronic charge display and information system) should be able to integrate the device 100 through standardized protocols.

Application in aviation:

one possible use scenario is to monitor the shoreline from the air. This technique (apparatus 100) can also be integrated in a helicopter by using an aeronautical compatible direction-finding system. Through the corresponding flying height and speed of the helicopter, the offshore communication monitoring can be realized in a wider range. The manufacturers of helicopter glass cockpit should also be able to integrate this application.

Further application:

support inland search and rescue organizations, e.g. monitoring coastal waters or organizing land rescue actions, e.g. coordinating police actions, emergency medical actions, fire rescue actions or actions of non-rescue organizations, e.g. mountain rescue.

Fig. 2 shows a block diagram of an apparatus 100 according to an embodiment of the invention. The apparatus 100 is configured to receive a voice radio signal 110, which may represent a voice signal (e.g., analog or digital), through a radio 230 (receiver). Thus, voice radio signal 110 may be transmitted by the subject and may be received by radio 230. Optionally, the apparatus 100 may comprise a radio 230, wherein the apparatus 100 may thus also be configured to transmit the voice radio signal 110 with the radio 230, while the actually transmitted voice radio signal 110 is further processed by the apparatus 100. According to embodiments, the radio 230 may be any radio or any source of voice signals (aeronautical radio band for aeronautical radio, waterborne radio band for waterborne radio, and/or emergency service radio for terrestrial radio).

According to an embodiment, the radio receiver of the radio 230 may send the voice radio signal 110 to the transcription unit 120 of the device 100, so that the device 100 may process the voice radio signal 110. The transcription unit 120 may be considered as an automatic transcription system of radio messages, wherein the transcription unit 120 is configured to convert the speech radio signal 110 into a text signal 112. To this end, the transcription unit 120 may include a speech identifier 124 (e.g., a message converted to text form (e.g., ASCII)) capable of converting the speech radio signal 110 into a text signal 112.

Furthermore, the transcription unit 120 may comprise, for example, a speech pattern identification 121, whereby the transcription unit 120 may be configured to extract a speech pattern code 122 from the speech radio signal 110 and provide it to the object determination unit 130 of the apparatus 100. The voice pattern code 122 may form a unique ID assigned to the radio message pattern by which the object from which the voice radio signal 110 originates may be identified. The object determination unit 130 may be identified by a voice mode code.

According to an embodiment, the transcription unit 120 is configured to convert the speech radio signal 110 into the text signal 112 using a neural network.

According to an embodiment, the apparatus 100 comprises an object localization unit 140 configured to determine position information 142 of an object from which the speech radio signal 110 originates. According to an embodiment, the object positioning unit 140 may comprise at least one radio direction finder 144₁To 144_n(e.g., part of a positioning device) or may be configured to communicate with at least one radio direction finder 144₁To 144_nCommunicating to determine direction finding data 142a₁To 142a_nAs the position information 142. Thus, the object-locating unit 140 may include n radio-direction finders 144₁To 144_nOr may be configured to interface with n radio direction finders 144₁To 144_nCommunication, wherein n represents a positive integer. Thus, the object-locating unit 140 may pass through the radio-direction finder 144₁To 144_nTo perform direction determination of radio signals, e.g., voice radio signals 110, with a plurality of direction finders 144₁To 144_nAllows for location determination of the wireless power supply. For example, if only one radio direction finder is used, only a rough area where a radio power source (object) is arranged can be determined as the position information 142. However, if multiple direction finders 144 are present and used₁To 144_nA very precise location of the wireless power supply can be determined by the object localization unit 140, for example by cross-location.

According to an embodiment, the object location unit 140 further may include or may be configured to communicate with a GPS receiver 144, an ADS-B receiver 146, an AIS receiver 147, a general location data receiver 148, and/or a compass 149 to receive location data, such as GPS data 142B₁ADS-B data 142B₂AIS data 142b₃And/or further generic location data 142b₄And 142b₅. Position data 142b₁To 142b₅May include the location of objects located within the area where the apparatus 100 determines with a certain probability the source of the voice wireless telecommunication 110. For example, the region may be represented by direction-finding data 142a₁To 142a_nAnd (4) generating. According to an embodiment, direction-finding data 142a₁To 142a_nCan be associated with the location data 142b₁To 142b₅Together forming location information 142 determined by the object-locating unit 140. Optionally, the target positioning unit 140 may further comprise a radar unit or may be configured to communicate therewith to receive further or alternative position data.

According to an embodiment, the GPS receiver 145 may be configured to determine the own position of the device 100. To this end, additionally or alternatively, a compass 149 may be used, wherein the compass 149 may determine its own heading, such as the heading of an object on which the apparatus 100 is disposed. Determining the position or heading of itself is advantageous in that the position of the object from which the speech radio signal 110 originates can be determined very quickly and efficiently and in relation to the position or direction of the device 100 or the object with the device 100.

According to an embodiment, the ADS-B receiver 146 may be configured to perform position determination of ADS-B transmitting objects, such as the position of aircraft in the environment. In accordance with an embodiment, AIS receiver 147 may be configured to perform a position determination of an AIS transmission object, such as the position of a ship in the environment. According to an embodiment, the universal position data receiver 148 may be configured to perform position determination and identification of any object, such as a land vehicle. Accordingly, the object locating unit 140 can locate the most diversified objects such as ships, airplanes, and/or land vehicles.

According to an embodiment, the location data 142b₁To 142b₅Which may be GPS position, course, speed and/or altitude relative to sea level.

Furthermore, the apparatus 100 comprises an object determination unit 130 configured to determine an object from which the speech radio signal 110 originates. According to an embodiment, the object determination unit 130 may also be referred to as automatic object identification with position determination. According to an embodiment, the object determination unit 130 receives the text signal 112 and/or the speech pattern code 120 from the transcription unit 120 and the location information 142, the location information 142 may comprise an area from which the speech radio signal originates as direction finding data 142a₁To 142a_nAnd may include location data 142b from the object location unit 140₁To 142b₅。

According to an embodiment, the object determination unit 130 may be divided into two processing units. The first processing unit 134 may be configured to perform general object recognition, such as ship recognition, aircraft recognition, and/or land vehicle recognition. Thus, the first processing unit 134 may process, for example, the location information 142. To this end, the object determination 130 may be configured to determine the location data 142b of the location information 142₁To 142b₅Direction finding data 142a associated with location information 142₁To 142a_nA comparison is made to determine the object from which the voice radio signal 110 having a particular probability of detection (which can be determined by the object determination unit 130) originated. Location information 142 includes, for example, a location or region from which voice radio signal 110 originated (e.g., direction-finding data 142 a)₁To 142a_n) And general location data 142b₁To 142b₅Which may include the location of all objects in the environment of the device 100. Thus, the object determination unit 130 may be configured to determine the location data 142b₁To 142b₅And direction finding data 142a₁To 142a_nAnd assigning a detection probability to the object determined in this way, wherein the detection probability depends on the match. In other words, the first processing unit 134 performs, for example, identification and position determination of an object (ship, airplane, or land vehicle) that transmits the radio signal 110 with a detection probability.

According to an embodiment, the detection probability may define the determined location information 142a₁To 142a_nWith the actual position 142b of the object₁To 142b₄To a degree of correspondence. Further or alternatively, the object determination unit 130 may be configured to determine the detection probability based on a probability of correct position information 142 of the object positioning unit 140, wherein correct may mean that the position data receiver 145, 146, 147, 148, 149 is comprised in determining the position data 142b₁To 142b₅Less than the lower limit of inaccuracy.

According to an embodiment, the object (e.g. a water vehicle, an aircraft or a land vehicle) detected in this way is sent by the first processing unit 134 to the second processing unit 136 of the object determination unit 130 together with the detection probability, the position, the route and/or further data. According to an embodiment, the object determination unit 130 may be configured to apply an algorithm for object data rendering to the detected object (e.g. by the first processing unit 134), e.g. by the second processing unit 136. By means of the algorithm, it is possible to combine all air, water and land vehicles on the one hand and vehicle information (location, route, etc.), radio message text 112, voice pattern codes 122, direction finding, etc. on the other hand, into one or more objects. In this way, the object determination unit 130 may be configured to, for example, determine the object having the highest detection probability as the object from which the voice radio signal 110 originates, thereby reducing all detected objects to one object. According to an embodiment, the object determination unit 130 is configured to determine the object identification of the object from the text signal 112, thereby reducing the detected plurality of objects to this one object. According to an embodiment, the object determination unit may be configured to determine the object based on the speech pattern code from which the speech radio signal 122 originates and thus reduce the detected object to this one object.

According to an embodiment, the object determination unit 130 may merge data of a plurality of objects when the plurality of voice radio signals 110 are processed simultaneously by the apparatus 100 or when an algorithm for object data rendering determines a plurality of objects considering transmitting the voice radio signals 110.

Furthermore, the apparatus 100 comprises an output unit 150 configured to assign the text signal 112 to an object and to provide the text signal. According to an embodiment, the output unit 150 may include an interface 152 for a data protocol and/or an internal graphics interface 154. Through the interface 152, the data determined by the apparatus 100 (e.g., text signals along with object identification and location and time) may be transmitted to an external device or external software to provide the data to the user of the apparatus 100. In this way, data may be transmitted, for example, to the ECDIS 153 and thus illustrated in an electronic chart. According to an embodiment, data is shown on a monitor 155 comprised by the apparatus 100 via the internal graphics interface 154.

According to an embodiment, the output unit 150 may be configured to assign at least two text signals 112 of the at least two speech radio signals 110 to corresponding objects and store them in a database (e.g. via the interface 152) and/or via a user interface of the device (e.g. the monitor 155).

In other words, fig. 2 shows an apparatus system and method for automatically placing handwritten voice messages (e.g., voice radio signals 110) transmitted via VHF maritime radio or aeronautical radio, i.e., explaining the same and optionally ensuring reliable sender distribution of each received voice message by connecting different information and communication technologies (AIS, ADS-B, GPS, and radio direction finding systems) onboard. Fig. 2 illustrates the system design in block diagram form.

According to an embodiment, a system (e.g., apparatus 100) is comprised of one or more computer systems and further data sources, e.g., processed as input data. As an output (e.g., output unit 150), the system has an internal graphical interface 154 adapted to illustrate on any monitor 155 the voice message (e.g., voice radio signal 110 as text signal 112) and the identified object. In addition, the system provides a data protocol interface 152 (e.g., NMEA) that can be processed by other information systems (e.g., ECDIS-electronic chart display information system 153) (see fig. 2).

For example, the following data or signals are processed as inputs (any combination of which is possible):

a) voice signal (e.g., voice radio signal 110) -a voice signal is, for example, an analog or digital signal that represents a received radio message and may be provided digitally by any radio 230 or intervening signal.

b) Direction finding data 142a₁-142a_n(Direction finders 1 to n 144)₁-144_n) -signal 142a₁-142a_nRepresenting direction finding data connected to the system via, for example, any protocol. Data 142a₁-142a_nIncluding, for example, direction finding, signal strength, adjusted frequency, and other data.

c) GPS data 142b₁GPS data 142b₁Of importance, for example for determining the position of itself (for example, own ships, aircraft, land vehicles, direction-finding stations in maritime traffic centres, direction-finding stations in airports). Further, data such as a change in UTC time and current position is optionally required.

d) ADS-B data 142B₂-ADS-B data 142B₂Typically obtained by the ADS-B receiver 146. These data include, for example, all relevant data of the aircraft, such as aircraft identification, ground position, altitude, speed, course and further data.

e) AIS data 142b₃With ADS-B data 142B₂Similarly, AIS data 142b₃Represents the location information of the vessel and is received, for example, by AIS receiver 147. Data 142b₃Other data such as ship identification, location, speed, route, etc. are also included.

f) The general location data receiver 148-the system should also be able to handle any data protocol by extension. Proprietary position determination and protocol systems may also be developed to extend the application of the system to other application areas (e.g., land, mountains, etc.).

g) Compass 149-the system optionally requires compass data to determine the direction of its own objects (e.g., its own ship, airplane, land vehicle, direction-finding station at sea traffic center, direction-finding station at airport). In general, compass data 142b₅Is assigned to the direction of the direction finding antenna. Thus, for example, each direction-finding antenna requires corresponding compass information. For a fixed direction-finding antenna, the direction may be directly input into the system.

According to an embodiment, the processing is performed in three steps. An analog or digital radio message (e.g., voice radio signal 110) is first converted to, for example, an ASCII text message 112 by an automatic transcription system 120. At the same time, the sender passes through one or more direction-finding systems 144, for example₁-144_nAnd (6) carrying out direction finding. By algorithms for automatic object identification (e.g. by the object determination unit 130) and location determination, for example, the sender of a radio message (e.g. an object) is identified and its location determined. For a detected object, for example, a corresponding detection or identification probability is indicated. Finally, the text message 112 of the radio message 110 is assigned to the corresponding object.

The detected object and the message 112 are output as output (e.g., output unit 150). The following options exist:

a) interface data protocol 152-the interface may be any defined interface or protocol that allows the system to be integrated into other systems.

b) Internal graphics interface 154-the system may also include its own proprietary graphical illustration of the determined data (its own graphical illustration on monitor/output location 155).

Fig. 3-5 below represent possible illustrations of a marine application-based graphical interface 154. Other applications, such as in the field of aeronautics or other applications, will follow the same logic.

Fig. 3 shows a graphical illustration of a transcribed radio message 112 (e.g., a text signal) of a vessel identified as a sender (100% identification probability 135 (e.g., detection probability)) in an electronic chart 153, according to an embodiment of the invention. Furthermore, an object or object identification 132 and optionally position information 142 are assigned to the radio message 112.

FIG. 4 shows three possible vessels (e.g., object 200) in the electronic chart 153₁To 200₃) Wherein the transcribed radio message 112 is assigned to an object 200 having an object identification 132, a graphical representation of the transcribed radio message 112 (e.g. a text signal)₁And has a highest identification probability 135 of 80% (detection probability)₁The location information 142. For the detected object 200₁To 200₃Of the invention can determine an identification probability 135₁To 135₃And assigns it to the corresponding object.

Fig. 5 shows a representation of an electronic chart 153 when the identity of the sender cannot be clearly determined by the device.

According to an embodiment, the inventive apparatus 100 may be arranged in a land station 300. By means of a radio direction finder, the apparatus may be configured to determine an area 220, in which area 220 the objects from which the speech radio signals originate are arranged with a probability of 135, 135₁To 135₃. According to the embodiments in fig. 3-5, the area 220 may be a signal beam determined by a radio direction finder.

Fig. 6 shows a block diagram of a method 1000 for processing a speech radio signal, wherein the method comprises converting the speech radio signal 1100 into a text signal by means of a transcription unit. Furthermore, the method 1000 comprises determining 1200, by the object determination unit, an object from which the speech radio signal originates. Further, the method 1000 comprises determining 1300, via the object localization unit, position information of an object from which the speech radio signal originates, assigning 1400 the text signal to the object, and providing 1500 the text signal assigned to the object by the output unit.

Although some aspects have been described in the context of an apparatus, it is clear that these aspects also represent a description of the respective method, such that a block or device of an apparatus also corresponds to a respective method step or a feature of a method step. Similarly, aspects described in the context of method steps also represent a description of the respective block or detail or feature of the respective apparatus. Some or all of the method steps may be performed by (or using) hardware devices, such as microprocessors, programmable computers, or electronic circuits. In some embodiments, some or several of the most important method steps may be performed by such an apparatus.

Embodiments of the invention may be implemented in hardware or software, depending on certain implementation requirements. Digital storage media, such as floppy disks, DVD, blu-ray discs, CD, ROM, PROM, EPROM, EEPROM or FLASH memories, hard drives or other magnetic or optical memories having electronically readable control signals stored thereon, which cooperate or are capable of cooperating with programmable computer systems to perform the respective methods, may be used. Accordingly, the digital storage medium may be computer-readable.

Some embodiments according to the invention comprise a data carrier comprising electronically readable control signals capable of cooperating with a programmable computer system so as to carry out one of the methods described herein.

Generally, embodiments of the invention can be implemented as a computer program product having a program code, which is operative for performing one of the methods when the computer program product runs on a computer.

For example, the program code may be stored on a machine-readable carrier.

Other embodiments include a computer program for performing one of the methods described herein, wherein the computer program is stored on a machine-readable carrier.

In other words, an embodiment of the inventive method is thus a computer program comprising a program code for performing one of the methods described herein, when the computer program runs on a computer.

A further embodiment of the inventive method is therefore a data carrier (or a digital storage medium or a computer-readable medium) having recorded thereon a computer program for executing one of the methods described herein. The data carrier, digital storage medium or computer readable medium is typically tangible or non-volatile.

A further embodiment of the inventive method is thus a data stream or a signal sequence representing a computer program for performing one of the methods described herein. The data stream or the signal sequence may for example be arranged to be transmitted via a data communication connection, for example via the internet.

Further embodiments include a processing device, such as a computer or programmable logic device, configured or adapted to perform one of the methods described herein.

A further embodiment comprises a computer having installed thereon a computer program for performing one of the methods described herein.

Further embodiments according to the present invention comprise an apparatus or system configured to transmit a computer program for performing at least one of the methods described herein to a receiver. For example, the transmission may be electronic or optical. For example, the receiver may be a computer, a mobile device, a memory device, or the like. For example, the apparatus or system may comprise a file server for transmitting the computer program to the receiver.

In some embodiments, a programmable logic device (e.g., a field programmable gate array, FPGA) may be used to perform some or all of the functions of the methods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor to perform one of the methods described herein. In general, the methods are preferably performed by any hardware device. This may be universally applicable hardware such as a Computer Processor (CPU) or hardware specific to this approach such as an ASIC.

The apparatus described herein may be implemented, for example, by a hardware device or a computer or a combination of a hardware device and a computer.

The apparatus described herein or any component of the apparatus described herein may be implemented at least partly in hardware and/or software (computer program).

The methods described herein may be implemented, for example, by a hardware device or a computer or a combination of a hardware device and a computer.

The methods described herein or any components of the methods described herein may be performed at least in part by hardware and/or software (computer programs).

Further embodiments of the invention will now be described.

A first embodiment provides an apparatus (100) for processing a voice radio signal (110), comprising:

a transcription unit (120) configured to convert a speech radio signal (110) into a text signal (112);

an object determination unit (130) configured to determine an object (200) from which the speech radio signal (110) originates;

an object localization unit (140) configured to determine position information (142) of an object (200) from which the speech radio signal (110) originates;

an output unit (150) configured to assign a text signal (112) to an object (200) and to provide the text signal.

A second embodiment provides an apparatus (100) according to the first embodiment, wherein the object localization unit (140) is configured to determine a region (220) in which the objects are arranged with a probability (200) as position information (142) and

wherein the object localization unit (140) comprises or is configured to communicate with at least one localization apparatus to determine a source of the speech radio signal as an area (220).

A third embodiment provides an apparatus (100) according to the second embodiment, wherein the positioning apparatus comprises at least one radio direction finder (144)₁To 144_n)。

A fourth embodiment provides an apparatus (100) according to the first to third embodiments, wherein the object localization unit (140) is further configured to receive position data (142 b) of an object (200)₁To 142b₅)。

A fifth embodiment provides an apparatus (100) according to the fourth embodiment, wherein the object localization unit (140) comprises an AIS receiver (147), an ADS-B receiver (146), a radar unitAnd/or a general location data receiver (148) or is configured to communicate therewith to receive location data (142 b)₁To 142b₅) And

wherein the position data (142 b)₁To 142b₅) Including GPS position, course, speed, and/or altitude relative to sea level.

A sixth embodiment provides an apparatus (100) according to one of the first to fifth embodiments, wherein the object determination unit (130) comprises or is configured to communicate with an AIS receiver (147), an ADS-B receiver (146) and/or a generic object identification receiver to obtain object identification data (132) of the at least one object (200), the position (210) of which at least partially matches the position information (142) determined by the object positioning unit (140).

A seventh embodiment provides an apparatus (100) according to the sixth embodiment, wherein the object identification data (132) comprises a calling number of a mobile services at sea (MMSI), an object name, a target of the object (200), a load of the object (200) and/or a size of the object (200).

An eighth embodiment provides the apparatus (100) according to the first to seventh embodiments, wherein the object determination unit (130) is configured to determine a detection probability (135 ) of at least one object (200) whose position (210) at least partially matches the determined position information (142)₁To 135₃) And an

Wherein the object determination unit (130) is configured to determine that it has the highest detection probability (135 )₁To 135₃) Is the object (200) from which the voice radio signal (110) originates.

A ninth embodiment provides an apparatus (100) according to the eighth embodiment, wherein the detection probabilities (135 )₁To 135₃) Defining a degree of correspondence of the determined position information (142) with an actual position (210) of the object (200), and/or

Wherein the object determination unit (130) is configured to determine the detection probability (135 ) based on a probability of correct position information (142) of the object positioning unit (140)₁To 135₃)。

A tenth embodiment provides an apparatus (100) according to one of the first to ninth embodiments, wherein the object determination unit (130) is configured to communicate with the transcription unit (120) to determine object identification data (132) of the object (200) from a text signal (112).

An eleventh embodiment provides an apparatus (100) according to one of the first to tenth embodiments, wherein the transcription unit (120) is configured to extract a speech pattern code (122) from the speech radio signal (110) and provide it to the object determination unit (130),

wherein the object determination unit (130) is configured to determine the object (200) from which the speech radio signal (110) originates based on the speech mode code (122).

A twelfth embodiment provides an apparatus (100) according to one of the first to eleventh embodiments, wherein the transcription unit (120) is configured to convert the speech radio signal (110) into a text signal (112) using a neural network.

A thirteenth embodiment provides an apparatus (100) according to the first to twelfth embodiments, the apparatus (100) being configured to process at least two voice radio signals (110) simultaneously and/or offset in time, and

wherein the output unit (150) is configured to assign at least two text signals (112) of the at least two speech radio signals (110) to corresponding objects (200) and to provide the same text signals (112) to the apparatus (100) chronologically via the user interface (155) and/or storing them in a database.

A fourteenth embodiment provides the apparatus (100) according to the first to thirteenth embodiments, wherein the output unit (150) is configured to provide the text signal (112), the assigned object (200), the location (210) of the object (200) and the input time of the voice radio signal to the apparatus (100) via the user interface (155) and/or storing it in a database.

A fifteenth embodiment provides the apparatus (100) according to the first to fourteenth embodiments, wherein the object (200) is a ship, an airplane, or a vehicle.

A sixteenth embodiment provides a method (1000) of processing a voice radio signal, the method comprising the steps of:

-converting (1100) the speech radio signal into a text signal by means of a transcription unit;

determining (1200), by an object determination unit, an object from which the speech radio signal originates;

determining (1300), by an object locating unit, position information of an object from which the speech radio signal originates; and

the text signal is assigned (1400) to the object and the text signal assigned to the object is provided (1500) via the output unit.

A seventeenth embodiment provides a computer program having a program code for performing the method according to the sixteenth embodiment, when the program is run on a computer.

The above-described embodiments are merely illustrative of the principles of the present invention. It will be understood that modifications and variations of the arrangements and details described herein will be apparent to those skilled in the art. It is the intention, therefore, to be limited only as indicated by the scope of the claims appended hereto rather than by the specific details presented by way of description and explanation of the embodiments herein.